permute_timewise_statistics {jlmerclusterperm}

Simulate cluster-mass statistics via bootstrapped permutations

Description

Simulate cluster-mass statistics via bootstrapped permutations

Usage

permute_timewise_statistics(
  jlmer_spec,
  family = c("gaussian", "binomial"),
  statistic = c("t", "chisq"),
  nsim = 100L,
  predictors = NULL,
  ...
)

Arguments

jlmer_spec

Data prepped for jlmer from make_jlmer_spec()

family

A GLM family. Currently supports "gaussian" and "binomial".

statistic

Test statistic for calculating cluster mass. Can be one of "t" (default) from the regression model output or "chisq" from a likelihood ratio test (takes about twice as long to calculate).

nsim

Number of simulations description

predictors

(Optional) a subset of predictors to test. Defaults to NULL which tests all predictors.

...

Optional arguments passed to Julia for model fitting. Defaults to fast = TRUE (when family = "binomial") and progress = FALSE.

Value

A simulation-by-time-by-predictor 3D array of cluster statistics, of class timewise_statistics.

See Also

make_jlmer_spec()

Examples

library(dplyr, warn.conflicts = FALSE)

# Specification object
spec <- make_jlmer_spec(
  weight ~ 1 + Diet, filter(ChickWeight, Time <= 20),
  subject = "Chick", time = "Time"
)
spec

# Simulation x Time x Predictor array of t-statistics from regression output
reset_rng_state()
null_statistics <- permute_timewise_statistics(spec, nsim = 3)
round(null_statistics, 2)

# Collect as dataframe with `tidy()`
permuted_timewise_stats_df <- tidy(null_statistics)
permuted_timewise_stats_df

# Permutations ran under the same RNG state are identical
reset_rng_state()
null_statistics2 <- permute_timewise_statistics(spec, nsim = 3)
identical(null_statistics, null_statistics2)

get_rng_state()
null_statistics3 <- permute_timewise_statistics(spec, nsim = 3)
identical(null_statistics, null_statistics3)